English   español  
Please use this identifier to cite or link to this item: http://hdl.handle.net/10261/142640
logo share SHARE   Add this article to your Mendeley library MendeleyBASE
Visualizar otros formatos: MARC | Dublin Core | RDF | ORE | MODS | METS | DIDL
Exportar a otros formatos:

Biología estructural de las ATPasas humanas RuvBL1 y RUVBL2 y su papel en remodelación de cromatina

AuthorsLópez-Perrote, Andrés
AdvisorLlorca, Óscar
KeywordsProteínas AAA+
Crio-microscopía electrónica y procesamiento de imagen
Issue Date2014
PublisherCSIC - Centro de Investigaciones Biológicas (CIB)
Universidad Complutense de Madrid
AbstractChromatin remodeling is a key event for the regulation of cellular processes such as transcription or DNA damage repair. Two main mechanisms to remodel chormatin have been describe, the modification of histones and specialized protein complexes that remodel genome packaging in an ATP-dependent manner, known as chromatin remodelers. There are four different families of remodelers described to date: SWI/SNF, ISWI, CHD and INO80. All of them use the energy of ATP hydrolysis to modify histone-DNA contacts in the nucleosomes, and they share a similar catalytic subunit belonging to the SWI2/SNF2 family of ATPases (Clapier & Cairns, 2009). The subunit composition of each remodeler is responsible for their different cellular functions. The INO80 chromatin-remodeling complex carries out the displacement of nucleosomes as well as replacement of histones to modify the DNA architecture, and it is involved in transcriptional activation and DNA repair (Conaway & Conaway, 2009). Two subunits of the INO80 complex are the highly conserved AAA+ ATPases RuvBL1 and RuvBL2, which also form part of a wide range of different macromolecular complexes. These proteins are involved in chromatin remodeling, transcriptional regulation, biogenesis of ribonucleoproteins and DNA repair, among other functions (Rosenbaum et al, 2013). In the INO80 complex, RuvBL1 and RuvBL2 subunits are essential for the recruitment of the Arp5 protein, resulting in the formation of the active remodeller complex (Jonsson et al, 2004). RuvBL1 and RuvBL2 assemble as large oligomers, including hexamers and dodecamers, but their involvement in the different macromolecular complexes of which they form part is not very well documented. There are some discrepancies about the oligomeric form of these proteins that is relevant in vivo, and their molecular mechanism is still unknown. I believe that the structural and biochemical study of these proteins is needed to improve our understanding on how these proteins functions in such a variety of complexes, thus resulting in their participation in a wide range of cellular events.
The Yin Yang 1 (YY1) transcription factor, that regulates transcription of many genes (Zhang et al, 2011), has been recently described as a novel subunit of the INO80 chromatin remodelling complex in mammalian cells (Cai et al, 2007; Wu et al, 2007). In this context, YY1 participates in transcription but also in genomic integrity maintenance. It seems possible that this protein, as part of the INO80 complex, might be directly involved in DNA repair through homologous recombination (Wu et al, 2007). Moreover, YY1 interacts in vivo and in vitro with different INO80 subunits, including the RuvBL1 and RuvBL2 proteins. In this thesis, we have explored the role of YY1 in homologous recombination in the context of INO80, and the involvement of the RuvBL1 and RuvBL2 proteins in this process. 2. Aim of this thesis The structural characterization of large macromolecular complexes is essential to understand its role in the cellular processes they act. These multi-protein assemblies frequently have a modular organization, comprising several funtional subcomplexes. The INO80 complex is involved in transcription (Shen et al, 2000), replication (Papamichos- Chronakis & Peterson, 2008), cell division, and DNA repair (Downs et al, 2004). In this thesis, we have addressed the biochemical and structural characterization of different subunits of the human INO80 chromatin-remodeling complex to investigate the structural and molecular bases that determine their functions.
To reach this aim, we have addressed the following specific objectives: 1. To express and purify full-length human RuvBL1 and RuvBL2 proteins, to characterize the oligomerization state of the complex they form and to solve its tridimensional structure. 2. To determine the molecular and structural mechanisms through which the RuvBL1- RuvBL2 complex regulates the biogenesis, maturation and remodeling of large macromolecular machines, including the INO80 complex, and to propose possible models of action. 3. To express, purify and characterize the structure of the human transcription factor YY1, as well as its nucleic acids binding properties. 4. To study the role of YY1 in homologous recombination in vivo, and to characterize the influence of the AAA+ ATPases RuvBL1 and RuvBL2 in this process in the context of the chromatin remodeler INO80. 3. Results and conclusions 3.1. Structural and biochemical characterization of the human RuvBL1-RuvBL2 complex. The full-length RuvBL1 and RuvBL2 proteins were coexpressed in bacterial cells to promote the assembly of their complex in vivo. Size exclusion chromatography (SEC) experiments revealed that the proteins assembled as hexamers and dodecamers. The sample was analyzed in the electron microscope, and images of complexes corresponding to the dodecamers were visualized. Interestingly, the analysis of the images suggested the presence of two different conformations of the dodecamer, or alternatively, two different complexes.
We obtained ab initio structures for both conformations, which we named compact and stretched, demonstrating the presence of two conformations of the complex. We also analyzed if this conformational transition could be due to a certain nucleotide binding state, by incubating the complex with different nucleotides and non hydrolysable nucleotide analogues and measuring the ratio between the compact and stretched conformations images. We did not found significant differences between the different conditions tested. Our interpretation is that the nucleotide binding was not affecting the conformation of the complex, but we could not exclude the possibility that the oligomer was not interchanging the nucleotide in our experiments, given that in the atomic structure of the RuvBL1 hexamer the nucleotide pocket seems to be blocked by the oligomerization of the protein (Matias et al, 2006). The structures for the compact and stretched RuvBL1-RuvBL2 complexes at 3 nm resolution were compatible with known atomic structures of the complex (Gorynia et al, 2011; Matias et al, 2006). Some authors have suggested that the dodecameric RuvBL1-RuvBL2 complex could be an artifact induced by the histidine tags included in the recombinant proteins. To address this possibility, we digested the histidine tags in the complex, and we analyzed the oligomerization of the wild type complex compared to the tagged one. In SEC experiments, removal of the tag promoted the complete disassembly of the dodecamer, detecting only hexameric forms. In contrast, native electrophoresis assays showed the presence of dodecamers and hexamers in both samples, regardless of the presence or absence of the histide tag. In addition, we also detected the dodecamers when the wild type complex was observed in the electron microcope. Together, these results suggested that the histidine tags were not inducing an artifact oligomerization, but they were responsible for the stabilization of the dodecameric complex, making it more resistant to disassembly by dilution.
We also analyzed the RuvBL1-RuvBL2 dodecamer by cryoelectron microscopy, and we obtained 1.6 nm resolution structures for both conformations. This allowed us to modell the the interface between the two rings by fitting atomic structures in the electron microscopy maps. The compact conformation showed closer contacts between the DII domains involved in inter-rings interactions, while in the stretched conformation these domains were pulled upwards and slightly rotated, thus extending the complex along the longitudinal axis. The DII domains in the stretched complex showed a loosser interface between the two rings. The DII domain is structurally similar to the ssDNA-binding domain in RPA; in addition, isolated DII domains from RuvBL1 have been shown to bind different nucleotides substrates (Matias et al, 2006). We found that the putative nucleic acids binding regions seem more exposed in the stretched conformation compared with the compact conformation. Hence, we propose that these conformational changes might regulate the interaction with nucleic acids or, alternatively, binding of a certain nucleic acid could promote one of the states, regulating the functions of the complex. In addition, a more general role of these DII domains as protein-binding modules within larger macromolecular complexes could be possible. This hypothesis agrees with the recently described structure of the yeast SWR1 chromatin remodeling complex, where a RuvBL1-RuvBL2 heterohexamer acts as a platform for connecting different multisubunit modules through the DII domains (Nguyen et al, 2013). Each DII domain is directly connected to the ATPase core, so it is reasonable to hypothesize that there is a link between the conformation of the AAA+ core and the DII domains. Conformational changes in DII could also regulate the ATPase activity and/or vice versa.
In this work, we have revealed the structure of the full-length human RuvBL1-RuvBL2 double-ring complex, and we he resolved the discrepancies for the different structures described. We show that RuvBL1-RuvBL2 displays two distinct conformations and propose that these transitions could have a functional impact in the context of the large macromolecular complexes containing RuvBL1-RuvBL2. These conformational transitions could be part of the mechanism of remodeling by converting the complex from one state to another, and all these changes could be somehow interconnected with a modulation of the ATPase activity and/or nucleic acids binding properties, thanks to the connection between the DII domains and the AAA+ core.
3.2. Role of YY1 and INO80 in homologous recombination A novel subunit of the INO80 chromatin-remodeling complex that we have studied is the YY1 transcription factor. This protein is involved in the transcriptional regulation of a wide range of genes (Gordon et al, 2006). An alternative role for YY1 has been recently described in homologous recombination (HR) in the context of INO80 and their RuvBL1-RuvBL2 subunits(Cai et al, 2007; Wu et al, 2007). We analyzed the participation of YY1 in homologous recombination by combining cellular, biochemical and structural approaches. We found that YY1 mediates the promotion of RAD51 filaments during homologous recombination in vivo, in cooperation with INO80 and RuvBL2. These findings corroborates previous results by Wu et al, but also revealed new data supporting a role of YY1 during HR. We also performed an in vitro characterization of YY1. Recombinant YY1 was able to interact with Holliday junctions DNA substrates, an intermediate in HR, and that this interaction promoted the assembly of several monomers of the protein in the nucleic acid. We explored the possibility of the oligomerization of YY1, and we found that the protein assembles two different oligomers in the absence of DNA. These YY1 oligomers, that we named complex A and complex B, were analyzed in the electron microscope. Complex A was solved as a “bell-shaped” dimeric core structure. The head region of the dimer supported major contacts of the structure, while the other segments assembled as arms slightly protruding around a central cavity. We obtained low resolution ab initio structures for the complex B, which showed mass and dimensions compatible with the assembly of two units of complex A. Despite several efforts, we were not able to refine these structures to higher resolution, due to the impossibility of assigning the axis of oligomerization. We proposed two alternatives forms of interaction for these complexes, a back-to-back contacts between two dimers of YY1, and a lateral association. We demonstrated that purified YY1 oligomers, mainly complex B, were able to bind Hollid junction intermediates in a sequence-independent manner. YY1 oligomers also interact with the RuvBL1-RuvBL2 complex in vitro. Binding of YY1 to Holliday junctions was improved when ATPases RuvBL1-RuvBL2 were present, suggesting that the 3 proteins are cooperating in binding HR intermediates, probably as a tripartite complex.
Together, these results suggest that YY1, in the context of INO80 and in cooperation with the RuvBL1-RuvBL2 subunits, is involved in promoting the formation of RAD51 filaments during HR. Moreover, the ability of YY1 to bind HR intermediates in cooperation with RuvBL1-RuvBL2 could be a reflect of its participation in additional steps during HR. Oligomerization of YY1 could have a biological significance, since these oligomers bind HR intermediates, and this could be involved in the molecular mechanism for maintaining genomic integrity.
Description202 p.-49 fig.-13 tab.
Appears in Collections:(CIB) Tesis
Files in This Item:
File Description SizeFormat 
Tesis_Andres_Lopez_Perrote_2014.pdf37,88 MBAdobe PDFThumbnail
Show full item record
Review this work

WARNING: Items in Digital.CSIC are protected by copyright, with all rights reserved, unless otherwise indicated.